[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#718533: linux-image-3.10-1-amd64: conf/modules typo "dm-raid45" makes RAID5 arrays unbootable



On Thu, 2013-08-01 at 16:35 -0700, Jesse Molina wrote:
> Package: initramfs-tools
> Version: 0.113
> Severity: serious
> Justification: unbootable
> 
> 
> I rebooted a system today on linux-image-3.10-1-amd64 and it was
> unable to boot.
> 
> linux-image-3.2.0-x-amd64 works fine.  A "initrd.img-3.2.0-4-amd64"
> file was generated at the same time the 3.10 one was crated and it
> boots normally.  I don't know what the difference is.
> 
> The problem is that the raid456 module is not being loaded by
> initramfs.  I manually loaded the module with "modprobe raid5",
> assembled by arrays with mdadm, and was able to boot normally.
> 
> The initramfs file /conf/modules says "dm-raid45", but I suspect it
> should say "dm-raid456".

Neither of these modules exists, nor did they exist in Linux 3.2.  Could
be your own configuration error or a bug in the dmraid hook (*not*
initramfs-tools itself).

[...]
> -- /proc/mdstat
> Personalities : [raid6] [raid5] [raid4] [raid0] [raid1] 
> md5 : active raid0 sda1[0] sde1[4] sdd1[3] sdc1[2] sdb1[1]
>       2441914880 blocks super 1.2 512k chunks
>       
> md3 : active raid5 sdf4[5] sdj4[4] sdi4[3] sdh4[2] sdg4[1]
>       1912129536 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
>       
> md2 : active raid5 sdf3[5] sdj3[4] sdi3[3] sdh3[2] sdg3[1]
>       39057408 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
>       
> md1 : active (auto-read-only) raid5 sdf2[5] sdj2[4] sdi2[3] sdh2[2] sdg2[1]
>       1947648 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
>       
> md0 : active raid5 sdf1[5] sdj1[4] sdi1[3] sdh1[2] sdg1[1]
>       382976 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
>       
> unused devices: <none>
[...]

And this indicates you are using md-RAID, so presumably you are not also
using dm-RAID...?

But I suspect that the boot failure probably has nothing at all to do
with this configuration error; see bug #678696.

Ben.

-- 
Ben Hutchings
This sentence contradicts itself - no actually it doesn't.

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: